mcp-server-semgrep/semgrep-rules/python/lang/security/audit/dangerous-code-run-audit.yaml

MCP Server for Semgrep Integration - static code analysis with AI

rules:
- id: dangerous-interactive-code-run-audit
  patterns:
  - pattern-either:
    - pattern: |
        $X.push($PAYLOAD,...)
    - pattern: |
        $X.runsource($PAYLOAD,...)
    - pattern: |
        $X.runcode(code.compile_command($PAYLOAD),...)
    - pattern: |
        $PL = code.compile_command($PAYLOAD,...)
        ...
        $X.runcode($PL,...)
  - pattern-either:
    - pattern-inside: |
        $X = code.InteractiveConsole(...)
        ...
    - pattern-inside: |
        $X = code.InteractiveInterpreter(...)
        ...
  - pattern-not: |
      $X.push("...",...)
  - pattern-not: |
      $X.runsource("...",...)
  - pattern-not: |
      $X.runcode(code.compile_command("..."),...)
  - pattern-not: |
      $PL = code.compile_command("...",...)
      ...
      $X.runcode($PL,...)
  message: >-
    Found dynamic content inside InteractiveConsole/InteractiveInterpreter method.
    This is dangerous if external data can reach this function call because it allows
    a malicious actor to run arbitrary Python code.
    Ensure no external data reaches here.
  metadata:
    cwe:
    - "CWE-95: Improper Neutralization of Directives in Dynamically Evaluated Code ('Eval Injection')"
    owasp:
    - A03:2021 - Injection
    references:
    - https://semgrep.dev/docs/cheat-sheets/python-command-injection/
    category: security
    technology:
    - python
    confidence: LOW
    subcategory:
    - audit
    likelihood: LOW
    impact: HIGH
  severity: WARNING
  languages:
  - python